INDEX

Explanations

references to software objects or identifiers within a computer program

oai_token-act-pair · gpt-3.5-turbo

references to "Object" and related programming concepts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

burg

-0.73

aukee

-0.73

 Deity

-0.70

tml

-0.70

bach

-0.70

ornia

-0.69

lights

-0.66

stakes

-0.65

good

-0.65

few

-0.64

POSITIVE LOGITS

ÑĮ

0.91

owicz

0.83

ives

0.81

dump

0.78

adesh

0.77

 Luthor

0.77

Lak

0.73

Obj

0.73

 Object

0.72

orate

0.72

Activations Density 0.025%

references to software objects or identifiers within a computer program

references to "Object" and related programming concepts

No Comments

No Known Activations